Expedited Surgery Increases Mortality Risk in Endometrial Cancer

David I. Shalowitz, MD

Patients with endometrial cancer who receive surgery in the first 2 weeks after diagnosis have worse survival outcomes, according to a study recently reported in the American Journal of Obstetrics and Gynecology.¹

Moreover, an extended delay in surgical treatment was also identified to be a risk factor for mortality in low-risk endometrial cancers and was likely associated with poor access to specialty care.

“We suspect that physicians diagnosing endometrial cancer may believe, not unreasonably, that the best thing they can do for those patients is to operate as soon as possible, because if they wait too long the cancer could progress, resulting in a worse outcome,” senior author David I. Shalowitz, MD, said in a press release.²

“But, the results of our study suggest that presurgical care and referring patients to a gynecologic oncologist may be more important,” added Shalowitz, who is a fellow in Gynecologic Oncology at the Perelman School of Medicine at the University of Pennsylvania.

The authors of the study suggested that the target interval between diagnosis and surgical treatment of endometrial cancers is ≤8 weeks, though they also stressed that both referral to a surgeon and preoperative optimization should be prioritized over expedited surgery.

The results of the study demonstrated that patients with low-risk cancers had a median survival time of 47.6 months (interquartile range [IQR], 25.8-73.6). The 5-year crude survival was highest when surgery was performed 3 weeks after diagnosis, and survival declined thereafter.

Patients who received surgical treatment during the first and second weeks following diagnosis (11.7%) had a higher risk of death (HR, 1.9; 95% CI, 1.7-2.1) compared with patients who underwent surgery in the third week after diagnosis (HR, 1.1; 95% CI, 1.1-1.2). Even after the researchers adjusted for age, stage, race, year of diagnosis, and additional clinical and health system characteristics, early surgery in the first and second weeks after diagnosis was still independently associated with a higher risk of death.

When surgery was performed 8 weeks postdiagnosis, the mortality risk was significantly higher than baseline and worsened as the interval between surgery and diagnosis increased.

For patients with high-risk cancers, the median survival time was 38.6 months (IQR, 18.9-67.1). As with patients with low-risk endometrial cancers, the 5-year crude survival was highest, again, when patients underwent surgery in the third week after diagnosis. Survival outcomes continued to decline thereafter.

High-risk patients who had surgery during the first and second weeks after cancer diagnosis (15.9%) had a hazard ratio for death of 2.1 (95% CI, 1.9-2.2). Surgery during this interval remained independently associated with death after the study authors adjusted for age, stage, race, year of diagnosis, and additional clinical and health system characteristics.

Surgery after the third week postdiagnosis was not correlated with a statistically significant increase in the adjusted risk of mortality, with the exception of an isolated increase seen in the nineteenth week after diagnosis.

The National Cancer Database provided cases of endometrial cancer between 2003 and 2012 for this study. Low-risk (grade 1 and grade 2 endometrioid histologies) and high-risk (grade 3 endometrioid and all other epithelial histologies) tumors were analyzed separately. The researchers limited the overall analysis to only cases for which there were data that surgery was the only modality pursued, or occurred before any hormonal, radiation, or chemotherapy treatments.

Given the results showing increased mortality risk accompanying surgery given in the first 2 weeks postdiagnosis, the investigators compared clinical and process-based factors for patients who had surgery in that short time period with patients who underwent surgery 3 and 4 weeks after diagnosis.

Patients with low-risk cancers who received surgical treatment in the first week after diagnosis were more likely to be very young or very elderly (<45 years or >85 years), Black, uninsured or with Medicaid insurance, have advanced stage disease, and undergo both diagnosis and treatment at the reporting Commission on Cancer (CoC)&shy;—accredited hospital. Plus, these patients were less likely to be treated at high-volume hospitals or be treated with lymphadenectomy.

Patients with high-risk endometrial cancers who had surgery in the first week after diagnosis were more likely to be elderly (>85), Black, uninsured or with Medicaid insurance, have advanced stage disease, and receive both diagnosis and treatment at the reporting CoC institution. Similar to the low-risk patients, they were also less likely to be treated at high-volume hospitals or undergo lymphadenectomy.

When compared with patients treated in the third or fourth week after diagnosis, 30-day postoperative mortality was significantly higher among patients treated in the first or second week post-diagnosis. For patients with low-risk cancers, this difference was 0.7% versus 0.4% (P <.001), and for patients with high-risk cancers, the difference was 2.5% versus 1.0% (P <.001).

For patients with low-risk endometrial cancers, the independent associations with added time-to-surgery of at least 1 week were observed with Black race (1.1 weeks; 95% CI, 0.9-1.4), lack of insurance (1.3 weeks; 95% CI, 1.1-1.5), Medicaid insurance (1.7 weeks; 95% CI, 1.5-1.9) and Charlson-Deyo comorbidity score >1 (1.0 weeks; 95% CI, 0.8-1.2).

For patients with high-risk cancers, independent associations with added time-to-surgery of at least 1 week were somewhat similar, as these correlations were observed with lack of insurance (1.4 weeks; 95% CI, 0.9-1.9) and Medicaid insurance (1.4 weeks; 95% CI, 1.1-1.7).

Disease stage was not consistently associated with the interval between diagnosis and surgery for patients with either low- or high-risk cancers.

Shalowitz DI, Epstein AJ, Buckingham L, et al. Survival implications of time to surgical treatment of endometrial cancers [published online December 8, 2016]. Am J Obstet Gynecol. doi:10.1016/j.ajog.2016.11.1050.